Despite the general success of autologous vein grafts and synthetic grafts (for large caliber vessels) as replacement conduits in vascular surgery, both the common lack of suitable autologous tissue (especially in the youngest and very oldest of patients) and the continuing overall high failure rates remain as significant limitations, especialy for small caliber replacements. There is, therefore, a pressing need for another strategy. Over the past few decades, tissue engineered vascular grafts (TEVGs) have advanced from benchtop to bed- side, with clinical trials now underway in both children and adults. These advances have arisen primarily via laborious trial-and-error comparisons of different biodegradable polymeric scaffolds defined by different surface chemistries, mechanical properties, and geometric characteristics (e.g., pore sizes, fiber diameters, and porosities). Pre-clinical studies have necessarily focused on safety and efficacy, which has primarily meant sufficient suture retention and burst pressure, thrombo-resistence, and the lack of formation of stenosis and aneurysm in vivo. Notwithstanding these many successes, there has yet to be a formal attempt to optimize scaffold design to yield biomechanical properties closer to native and having long-term biological stability. Given the continued advances in fabrication techniques, an almost limitless combination of scaffold parameters is now possible. It is inconceivable, however, that trial-and-error comparisons can possibly identify an optimal combination. Hence, we suggest a new paradigm for polymeric scaffold design - we will meld concepts of nondimensionalization, parameter sensitivity, and optimization within a novel validated computational model of in vivo neovessel development with 3 proven mouse models to identify and test a new optimal scaffold design. Toward this end, we will seek a bilayered design consisting of an inner porous layer of compliant poly(glycerol sebacate) that encourages cellular infiltration and an outer less porous, stiffer poly(e-caprolactone) sheath that supports the inner layer during its rapid degradation and replacement with neotissue. The computational model will be informed and refined via a small series of initial experiments that reveal in vivo the effects o extremes in porosity, fiber diameter, and stiffness while delineating overlapping roles of inflammatory- and mechano- mediated matrix production. Using formal concepts of nondimensionalization, parameter sensitivity, uncertainty quantification, and optimization, we then will identify via thousands of simulations those few scaffold designs that are most promising. These designs will be fabricated as new scaffolds and tested in vivo in mice for up to 2 years. Comparisons with results from the first series of tests will reveal if the predicted desigs are indeed significantly better; if not, we can iterate the process. Successful completion of our aims will establish a new computational-experimental paradigm for scaffold design, result in a much improved TEVG, and serve as an archetype for other tissue engineering applications via novel experimental methods, biomaterials, & modeling.